论文部分内容阅读
振动模态分析和模态参数识别是动态测试的一个重要研究方向。模态参数在模型的修正、响应的预测、系统的健康检测及控制等方面有着重要的作用。但动态测试的不确定度分析,尤其是模态参数的不确定度的研究还十分缺乏。本文主要基于贝叶斯方法,通过傅立叶变换(FFT)建立时域数据和频域数据之间的对应关系。根据共振频带内的多个模态的响应数据得到相对应的模态参数,优化得到模态参数的最佳估计值,评定模态参数识别的不确定度。在固支梁的模态实验中,加速度传感器采集环境激励中的振动数据,运用贝叶斯法进行处理得到模态参数的最佳估计值。在此基础上,通过模态参数的最佳估计值,以及仪器的检定报告数据,结合合成不确度分析方法,系统分析了模态参数识别的不确定度。
Vibration modal analysis and modal parameter identification are important research directions in dynamic testing. Modal parameters play an important role in model modification, response prediction, system health monitoring and control. However, the uncertainty analysis of dynamic testing, especially the research on the uncertainty of modal parameters, is still lacking. This paper mainly based on Bayesian method, through the Fourier transform (FFT) to establish the correspondence between time domain data and frequency domain data. The corresponding modal parameters are obtained according to the response data of multiple modalities in the resonance frequency band, and the best estimate of the modal parameters is optimized to evaluate the uncertainty of modal parameter identification. In the mode experiment of the fixed beam, the accelerometer collects the vibration data in the environment excitation, and uses the Bayesian method to get the best estimate of the modal parameter. Based on this, the uncertainty of modal parameter identification was systematically analyzed by the best estimate of modal parameters and the test report data of the instrument, combined with the synthetic uncertainty analysis method.